Aneurysmal subarachnoid hemorrhage (SAH) afflicts 30,000 patients/year in the U.S.A. Amongst those who survive the initial bleed, 15% die or suffer disabilities because of Delayed Ischemic Deficit (DID) due to vasospasm. Enormous research efforts into the pharmacological prevention and/or treatment of vasospasm have resulted in only minimal improvement for this problem. In a dog model of vasospasm, transluminal balloon angioplasty (TBA) immediately after clot placement was able to totally prevent vasospasm. In a preliminary study of 12 patients with severe bleeds (Fisher grade III, and III + IV), TBA performed within three days frm their SAH also totally eliminated DID and transcranial doppler (TCD) defined moderate to severe vasospasm, although the expected incidences were 45% and 80%, respectively. However, there was one procedure-related death due to patient movement of a non-intubated, non- paralyzed/anesthetized patient. We now propose to do a multi-center, randomized trial of standard treatment versus standard treatment plus prophylactic TBA. In three centers over 2.5 years, we will enroll 185 patients with severe bleeds (Fisher grade III) and 120 patients with less severe SAH (Fisher grade I, II) in this study and randomize them separately. In the severe group, TBA will be performed on all vessels forming the Circle of Willis, in the less severe group, only the aneurysm-bearing part of the circulation. TBA will be performed immediately following surgery (or endovascular treatment) under general endotracheal anesthesia and/or paralysis to maximize safety. Primary outcome will be the dichotomized Glasgow Outcome Scale (GOS) at three months. In the severe group, the number of patients is sufficient to detect an increase in favorable outcome from 40 to 70%. In the less severe group, we will have enough data to determine whether a larger multi- center trial should be done to establish the role of prophylactic TBA. Secondary outcome measures are the full five-point GOS, procedure- related death, TCD data, the occurrence of DID, and economic data. Finally 50 years after the first description of vasospasm, it appears that an effective preventative treatment is ready to be tested.